Bacterial community assembly and turnover within the intestines of developing zebrafish

Abstract

Background: The majority of animal associated microorganisms are present in digestive tract communities. These intestinal communities arise from selective pressures of the gut habitats as well as host's genotype are regarded as an extra 'organ' regulate functions that have not evolved wholly on the host. They are functionally essential in providing nourishment, regulating epithelial development, and influencing immunity in the vertebrate host. As vertebrates are born free of microorganisms, what is poorly understood is how intestinal bacterial communities assemble and develop in conjunction with the development of the host.

Methodology/principal findings: Set within an ecological framework, we investigated the bacterial community assembly and turnover within the intestinal habitats of developing zebrafish (from larvae to adult animals). Spatial and temporal species-richness relationships and Mantel and partial Mantel tests revealed that turnover was low and that richness and composition was best predicted by time and not intestinal volume (habitat size) or changes in food diet. We also observed that bacterial communities within the zebrafish intestines were deterministically assembled (reflected by the observed low turnover) switching to stochastic assembly in the later stages of zebrafish development.

Conclusions/significance: This study is of importance as it provides a novel insight into how intestinal bacterial communities assemble in tandem with the host's development (from early to adult stages). It is our hope that by studying intestinal microbiota of this vertebrate model with such or some more refined approaches in the future could well provide ecological insights for clinical benefit. In addition, this study also adds to our still fledgling knowledge of how spatial and temporal species-richness relationships are shaped and provides further mounting evidence that bacterial community assembly and dynamics are shaped by both deterministic and stochastic considerations.

Figure 1. Bacterial richness relationships with (A) habitat size (volume nL) and (B) time (days post-fertilization) plotted on log₁₀ scale axes.

Given are the (A) species-volume relationship and (B) species-time relationship power law equations. For (A): r ² = 0.25, F _1,28 = 9.16, P<0.005; and (B) r ² = 0.26, F _1,28 = 10.0, P<0.004. Also given are the 95% confidence and prediction intervals (inner and outer dashed lines, respectively).

Figure 2. Changes in bacterial community similarity within and between time points.

Bars indicate within time point similarity and circles represent similarity between adjacent time points. Similarity is measured by the Sørensen index of similarity (S _SOR). Error bars represent the standard deviation of the mean.

Figure 3. A comparison of changes in similarity values for within and between time points.

Bars indicate within time point similarity and circles represent similarity between adjacent time points. Similarity is measured by the Raup and Crick (S _RC) probability-based index of similarity. 0.95>S _RC>0.05 similarity is no greater than expected by chance, S _RC<0.05 significant dissimilarity, S _RC>0.95 significant similarity. Error bars represent the standard deviation of the mean.